Evaluation of an Innovative Point-of-Care Rapid Diagnostic Test for the Identification of Imported Malaria Parasites in China

Background: China was certified malaria-free by the World Health Organization on 30 June 2021. However, due to imported malaria, maintaining a malaria-free status in China is an ongoing challenge. There are critical gaps in the detection of imported malaria through the currently available tools, especially for non-falciparum malaria. In the study, a novel point-of-care Rapid Diagnostic Test designed for the detection of imported malaria infections was evaluated in the field. Methods: Suspected imported malaria cases reported from Guangxi and Anhui Provinces of China during 2018–2019 were enrolled to evaluate the novel RDTs. Diagnostic performance of the novel RDTs was evaluated based on its sensitivity, specificity, positive and negative predictive values, and Cohen’s kappa coefficient, using polymerase chain reaction as the gold standard. The Additive and absolute Net Reclassification Index were calculated to compare the diagnostic performance between the novel RDTs and Wondfo RDTs (control group). Results: A total of 602 samples were tested using the novel RDTs. Compared to the results of PCR, the novel RDTs presented sensitivity, specificity, PPV, NPV, and diagnostic accuracy rates of 78.37%, 95.05%, 94.70%, 79.59%, and 86.21%, respectively. Among the positive samples, the novel RDTs found 87.01%, 71.31%, 81.82%, and 61.54% of P. falciparum, P. ovale, P. vivax, and P. malariae, respectively. The ability to detect non-falciparum malaria did not differ significantly between the novel and Wondfo RDTs (control group). However, Wondfo RDTs can detect more P. falciparum cases than the novel RDTs (96.10% vs. 87.01%, p < 0.001). After the introduction of the novel RDTs, the value of the additive and absolute Net Reclassification Index is 1.83% and 1.33%, respectively. Conclusions: The novel RDTs demonstrated the ability to distinguish P. ovale and P. malariae from P. vivax which may help to improve the malaria post-elimination surveillance tools in China.


Introduction
Malaria remains a serious public health problem worldwide and is caused by Plasmodium parasites. It is reported that malaria cases were still on the rise between 2020 and 2021. How-for malaria, with major outbreaks occurring in 1954, 1963, and 1971. Throughout history, malaria in Guangxi was mainly caused by P. vivax and P. falciparum. After the founding of New China, after more than 70 years of prevention and control, the incidence rate dropped from 296.7/10,000 in 1954 to less than 1/10,000 in 1987; The last local infection case was reported in 2012, and since 2013 there have been no local infection cases reported in the province. All cases occurred as imported cases, mainly from African and Southeast Asian countries and regions. Among them, a total of 3195 malaria cases were reported in Guangxi from 2010 to 2019, with the main species being P. falciparum, while the other three species and mixed infections were all reported as well [13,14].
In China, when suspected malaria patients presenting with symptoms of malaria, especially combined with a history of travel to a malaria-endemic area, seek medical care, the physician provides a diagnostic test for malaria, commonly microscopy or rapid diagnostic test (RDTs) [15]. However, due to their morphological similarity, P. ovale is easily and commonly misdiagnosed as an infection of P. vivax [16,17], which may lead to an inappropriate case management treatment response. For someone self-diagnosing for malaria, RDTs are available in Chinese pharmacies; however, these also cannot differentiate between P. malariae, P. ovale, and P. vivax infections. Thus, prompt and precise diagnostic tools that can detect and differentiate non-falciparum malaria species are needed.
In this study, a novel point-of-care RDT was designed for the detection of imported malaria infections and was evaluated using polymerase chain reaction (PCR) as the gold standard. We believe that the newly designed diagnostic tool can benefit the prevention of malaria re-establishment in the future.

Study Setting, Participants and Design
The provinces of Anhui and the Guangxi Autonomous Region were selected as study areas for the evaluation of the novel RDTs ( Figure 1). Historically, malaria was highly prevalent in Anhui province [12], which presents a high risk of re-establishment of malaria. Guangxi is a border province in southern China that exports large numbers of migrant workers to Africa. Since 2013, the number of imported malaria cases in Guangxi Province has been among the highest in China [5]. In China, each suspected malaria case should be mandatorily reported through the China Information System for Disease Control and Prevention (CISDCP) [18]. This is a real-world study based on the surveillance system of malaria in China. All suspected imported malaria patients reported in Anhui and Guangxi provinces from 2018 to 2019 were enrolled as participants in the study. Individuals were contacted by telephone to obtain verbal informed consent. An imported case was defined as a malaria infection acquired outside the country (in this study, China). In China, each suspected malaria case should be mandatorily reported through the China Information System for Disease Control and Prevention (CISDCP) [18]. This is a realworld study based on the surveillance system of malaria in China. All suspected imported malaria patients reported in Anhui and Guangxi provinces from 2018 to 2019 were enrolled as participants in the study. Individuals were contacted by telephone to obtain verbal informed consent. An imported case was defined as a malaria infection acquired outside the country (in this study, China).
In China, when a suspected case is reported through the network, the blood samples, including whole blood and smears that were collected from the patient before anti-malarial treatment [19], are sent to the provincial reference laboratory for final confirmation using microscopic examination and polymerase chain (PCR) reaction according to the malaria diagnostic criteria in China [20]. In this study, the real-time PCR method, used in the form of commercial real-time PCR Kits (Shanghai ZJ Bio-tech Co., Ltd., Shanghai, China), was taken as the gold standard to detect the malaria infection and further distinguish between Plasmodium species. The kits, targeting the 18s rRNA gene, were designed by referring to a previous study and provided internal control [21]. Before PCR test, DNA was extracted using the QIAamp DNA Mini kit (QIAGEN Inc., Hilden, Germany) in accordance with the manufacturer's instructions. Then, PCR was performed in a 40.4-µL reaction mixture containing 35 µL reaction mix, 0.4 µL enzyme mix, 1 µL internal control, and 4 µL DNA template. The reaction conditions were as follows: 37 • C for 2 min and 94 • C for 2 min, followed by 40 cycles at 93 • C for 15 s and 60 • C for 60 s. On the other hand, a commercial test strip (Diagnostic Kit for Malaria, Guangzhou Wondfo Biotech Co., Ltd., Guangzhou, China) detecting Pf-HRP2 (Human histidine rich protein 2) and Pan-lactate dehydrogenase (LDH) was taken as control in comparison to the novel RDTs in the laboratory setting. An imported case was defined as a malaria infection acquired outside the country (in this study, China).

Interpretation of the Results for RDTs
A novel malaria RDT was designed by the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention and tested in this study. It is not yet officially available commercially. The novel RDT ( Figure 2) is an immunochromatographic test strip, and has one control line and three test lines ("T1", "T2", and "T3"), detecting Pf-HRP2, Pv-speciifc LDH, and Pan-LDH, respectively. If the infection was caused by P. vivax, T2 and T3 line were simultaneously positive, whereas, if the infections were P. ovale and/or P. malariae, only the "T3" line was positive, with a negative "T2" line. Using the combination of "T2" and "T3" test lines, the novel mRDT can distinguish P. vivax from P. ovale and/or P. malariae (Table 1). Wondfo RDTs have one control line ("C") and two detection lines ("T1" and "T2"). Additionally, the T1 and T2 lines indicate P. falciparum and Plasmodium infections, respectively. Blood samples from participants were tested simultaneously with novel and Wondfo RDTs in the provincial laboratory reference as directed by the manufacturer. ed. Infect. Dis. 2023, 8,296 were P. ovale and/or P. malariae, only the "T3" line was positive, with a neg Using the combination of "T2" and "T3" test lines, the novel mRDT ca vivax from P. ovale and/or P. malariae (Table 1). Wondfo RDTs have one co and two detection lines ("T1" and "T2"). Additionally, the T1 and T2 line ciparum and Plasmodium infections, respectively. Blood samples from pa tested simultaneously with novel and Wondfo RDTs in the provincial labo as directed by the manufacturer.

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The Positive result for P. ovale and/or P. malariae

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The

Data Analysis
The categorical data are presented as percentages. Values are presented as the mean ± standard deviation for data that were normally distributed. Differences in proportions were compared using McNemar's χ2 test. Taking the results of PCR as the gold standard, the diagnostic performances of the novel and Wondfo RDTs were presented with the following parameters: sensitivity, specificity, positive (PPV) and negative predictive values (NPV), and Cohen's kappa coefficient, with their respective 95% confidence intervals. The formula for calculating PPV and NPV is: PPV = (true positives)/(true positives + false positives), NPV = (true negatives)/(true negatives + false negatives), respectively. Additive Net Reclassification Index (NRI) and absolute NRI are calculated to compare diagnostic performance between the Novel and Wondfo RDTs [22]. All statistical tests were two-sided, and p < 0.05 was considered statistically significant. The study data were recorded and entered into an Excel database (Microsoft Corporation, Redmond, WA, USA), and analysis was performed using SPSS 26.0 statistical software (SPSS Inc., Chicago, IL, USA). The thematic map of geographic distribution was created by MapInfo 15.0 (Pitney Bowes Inc., Troy, NY, USA).

Additive NRI and Absolute NRI
The additive NRI and absolute NRI were calculated to assess the improvement due to the novel RDTs introduced in the field, compared to Wondfo RDTs. The values of the additive NRI and absolute NRI are 1.83% and 1.33%, respectively. The results showed that there was no difference in diagnostic ability between the Novel and Wondfo RDTs (all p > 0.05) ( Table 4).

Discussion
The last indigenous malaria case in China was reported in 2016 and local transmission has been interrupted since 2017 [23]. China was certified malaria-free by the WHO in 2021 and is facing continued challenges due to imported malaria, particularly among male workers visiting Africa. Therefore, performing and sustaining a sensitive surveillance system that can detect suspected malaria cases in a prompt and accurate manner is the key. However, the inability to properly detect and distinguish malaria parasites is a huge barrier [24]. In field practice, microscopy and RDTs are common diagnosis methods for malaria in the health care setting in China. However, sustaining microscopy competency is extremely difficult due to the limited accumulation of experience [25]. RDT has the characteristics of easy operation and intuitive reading; therefore, it is the diagnostic method for malaria that has been recommended by the WHO. Further, RDTs, a vital supplement, extend access to diagnostic tools in areas where microscopy cannot be reliably maintained. RDT has been recommended to provide parasite diagnosis for suspected malaria cases by WHO [26].
For malaria detection, RDT that can distinguish between the types of malaria parasites, live and dead infections, and the sexual stage of parasitemia will be very helpful for diagnosis and guide intervention measures. The main challenge is in the field of lowlevel parasitemia. The examination of the life cycle stage of malaria parasite infection has identified many key targets, including the HRP2 protein (P. falciparum), parasite lactate dehydrogenase (pLDH, Plasmodium genus), and malaria parasite aldolase (Plasmodium genus). Therefore, distinguishing between P. falciparum and other species is not a simple task [27]. Piper conducted research on how existing combinations of pLDH antibodies perform in the differential diagnosis of P. falciparum, pan specificity, and malaria parasites, showing that differences in reactivity may be related to small differences on the surface of pLDH, with subtle amino acid changes being the cause of species specificity [28].
There is evidence to suggest that, in countries with low malaria transmission, due to the long-term absence of malaria cases, the awareness and vigilance of health systems and the preparedness of health workers towards the correct management of suspected malaria will decrease [29].
In China, infections caused by P. falciparum, P. vivax, P. malariae, and P. ovale have still been reported for many years. The malaria surveillance system needs to introduce RDTs with the ability to detect four species. However, thus far, only two Pf/Pan tests (Wondfo and BinaxNOW ® Malaria) have been registered in the National Medical Products Administration that could be used in health facilities. Wondfo RDTs are used more in the market for price reasons, and presented a better performance for detecting P. ovale compared to CareStart pLDH PAN and SD BIOLINE Pf/Pan RDTs [30]. According to the manufacturer's instructions, Wonfo RDTs can distinguish P. falciparum from non-falciparum species. However, they could not further differentiate non-falciparum species. P. vivax, a common species in all malaria-endemic areas, is distributed nationwide and is the main species related to the risk of malaria reintroduction. Due to their morphological similarity, P. ovale is easily and commonly misdiagnosed as an infection of P. vivax in the field, further leading to inappropriate interventions. A novel RDT has been designed to fulfil the gap in field practice.
Related studies have shown that, although the protein sequence of pLDH is very conservative within the same species, there are certain differences in the protein sequence of pLDH among the four human malaria parasites. Usually, the detection antibodies in RDT are monoclonal antibodies that are prepared based on a specific pLDH antigen of a certain insect species. Therefore, when detecting unknown samples, if the patient is infected with other insect species in the body, the detection antibodies in RDT cannot specifically bind to the antigen in the sample, resulting in false negative test results. The specificity experiment results showed that the monoclonal antibody of PfLDH only reacted with the samples of P. falciparum and did not react with the samples of P. vivax (P. vi-vax); similarly, PvLDH antibodies only react with P. vivax samples and do not react with P. falciparum samples [31].
In this study, the diagnostic performance of the novel RDTs was assessed using blood samples collected from the field. The novel RDTs found 87.01% of P. falciparum, 71.31% of P. ovale, 81.82% of P. vivax, and 61.54% of P. malariae infections. The diagnostic performance of the novel RDTs for non-falciparum species was similar to that of the Wondfo RDTs. Importantly, the novel RDTs had the ability to distinguish P. ovale and P. malariae infections from P. vivax. Although improvements are still needed to improve the novel RDTs, we believe that significant progress has been made in this initial development and through this diagnostic evaluation.
Although the diagnostic sensitivity of the novel RDTs for P. falciparum was lower than for Wondfo RDTs (87.01% vs. 96.10%), P. falciparum is the dominant species of imported malaria [4]. Novel RDTs need to fulfil the gap in future practical applications. The value of additive NRI and absolute NRI are 1.83% and 1.33%, respectively, which means that there are no obvious additional benefits to the introduction of the novel RDTs in the field. P. ovale and P. malariae were commonly considered as the 'bashful' malaria parasites, due to their low prevalence and limited geographic distribution [32]. However, the introduction of more sensitive molecular methods has provided more evidence that their geographic distribution is larger than previously speculated [33,34]. Further, scientific knowledge about the two species is very limited compared to P. falciparum and P. vivax. Developing diagnostic tools that target the two main species of Plasmodium is challenging. One factor significantly impacting its success is that parasitaemia is typically very low in infections caused by P. ovale and P. malariae (sub-microscopic malaria infections). According to a previous study, P. malariae only invades aged red blood cells (0.1% parasitaemia) and P. ovale preferentially invades youthful red blood cells (1% parasitaemia) [35]. This implies that the concentration of specific antigens targeted by novel RDTs may be lower than the threshold value. For example, in our field assessment, novel RDTs only found 61.54% infections of P. malariae.
Our study has two main limitations. First, limitations around clinical and patient-level RDTs occur frequently in non-endemic settings, especially in the presence of treatment delays [36]. Therefore, the performance of the novel RDTs may be underestimated. Second, the number of P. malariae and P. vivax cases is small.

Conclusions
The diagnostic power of novel RDTs for non-falciparum species detection was comparable to the Wondfo RDTs, and demonstrated the ability to distinguish P. ovale and P. malariae from P. vivax. We believe that the newly designed RDTs can benefit POR practices within China and other malaria-eliminating and POR countries with further improvements.